PSI - Issue 28

1598 M.Z. Sadeghi et al. / Procedia Structural Integrity 28 (2020) 1590–1600 M.Z. Sadeghi et al./ Structural Integrity Procedia 00 (2020) 000–000 to 0.706 mm on the other end of the joint. The overlap tolerance for the joint was considered to be ± 0.04 mm. Temperature tolerance of ±10 0 C was also considered to be included in the interaction between the influential parameters of the strain measurement. Since arranging the whole simulations by using 3D FE model was very time consuming, the simulation was carried out in 2D. Following procedures were considered for the simulations arranged in this part:  Although Strain Gauge Misalignment is one of the parameters considered for variation, it does not need a separate simulation as its effect may be measured by extracting the longitudinal strain values at the misalignment points.  Three different states – increase (+), decrease (–) and reference (0) and 3 variables, excluding SG for the reason stated above were designed in the simulations. Therefore, the total simulations required were 3 3 = 27 cases.  In the reference simulation, it was found that the longitudinal strain remains zero till 70-80% of peak force. Subsequently it undergoes deformation from that point. The influence of strain gauge misalignment on the measured strains of the so-called ZSP is shown in Fig. 10. This figure which provides the measured longitudinal strains versus normalized force, provides the experimental results as well. As it can be seen, a small deviation from the considered point on the adherent (ZSP) creates a big deviation in the measured strains. However, the backface strain misalignment towards right and left of the ZSP shows different trends. The deviation towards right (towards the side in which the load applied) pushes the reference curve to upwards whereas deviation to the left (close to the clamped side of the SLJ) shifts the reference curves downwards. The comparative results of the interaction of the influential parameters on the measured longitudinal strain is shown in Fig. 11. The combined interaction of overlap and adhesive thickness tolerances as well as temperature variation with and without strain gauge misalignment is shown in the figure. The experimental results shows better agreement with the upper and lower band created by interaction of geometric tolerances, temperature changes and strain gauge misalignment. Considering the geometric tolerances existing the specimens tested in the present work, the overlap length tolerance showed the lowest influence of the variation of blackface strain at ZSP. As a whole, strain gauge misalignment shows the highest influence on the variation of the backface strain measurement. After that the highest influential factors were respectively: the temperature variation, adhesive layer thickness and overlap length tolerances. It is interesting to know that all the factors mentioned above can only influence the maximum fracture load of the joint by less than 1%. However, they can have high influence on the measured backface strain.

Fig. 10. Influence of strain gauge misalignment versus normalaized force compared with FE.